In a recent edition of this journal Mikael Bask and Maria Melkersson suggested that moist snuff, snus to the Swedes, is not an effective tool in cigarette smoking cessation. The sole basis for this conclusion is their empirical findings concerning the cross-price elasticity of demand between cigarettes and snus. Specifically, they find that, “The cross-price elasticities are negative, which indicates that snus contributes to increased smoking. Thus, even if snus taking is less harmful than cigarette smoking, it is not advisable to encourage its use in smoking cessation programs.” We find this conclusion to be unwarranted. It is based on a model estimated with inappropriate data; it is based on a misspecified model with questionable theoretical underpinnings; and no evidence is presented as to the statistical significance of the cross-price effects. Furthermore, even if we ignore these technical criticisms, their conclusions are unfounded because they are based on an inappropriate conceptual experiment. We first address some technical problems with the Bask and Melkersson analysis, and then we address the fundamental flaw in their reasoning
In a recent edition of this journal Mikael Bask and Maria Melkersson suggested that moist snuff, snus to the Swedes, is not an effective tool in cigarette smoking cessation. The sole basis for this conclusion is their empirical findings concerning the cross-price elasticity of demand between cigarettes and snus. Specifically, they find that, “The cross-price elasticities are negative, which indicates that snus contributes to increased smoking. Thus, even if snus taking is less harmful than cigarette smoking, it is not advisable to encourage its use in smoking cessation programs.” We find this conclusion to be unwarranted. It is based on a model estimated with inappropriate data; it is based on a misspecified model with questionable theoretical underpinnings; and no evidence is presented as to the statistical significance of the cross-price effects. Furthermore, even if we ignore these technical criticisms, their conclusions are unfounded because they are based on an inappropriate conceptual experiment. We first address some technical problems with the Bask and Melkersson analysis, and then we address the fundamental flaw in their reasoning
A significant set of the public-choice literature considers the relationship between primary types and the resulting ideology of the winning candidate in primary and general elections. Research by Buel and Jackson [1991], for instance, suggests that primary voters with ideologically extreme views are more likely to participate in political activities than other voters. Gerber and Morton [1998] argue that the relevant median voter in closed primaries is more extreme than the median voter in general elections and that this difference is less pronounced in states with open or blanket primaries. In this paper, we build upon Gerber and Morton [1998] by adding a new measure of deviation from the median voter that explains why politicians may prefer one primary type to another. We suggest that institutional arrangements such as primary types, in addition to logrolling and signaling, may account for candidate deviation from the median voter. In the process, we re-evaluate this principal-agent relationship with a focus on the role of agent deviation under different primary types.
This article presents a simple model of rent control in which the costs and benefits can be analyzed in terms of Hicksian consumer and producer surpluses. Data from New York City are then used to examine the effect of the long-term rent control in that city. Estimates are presented which suggest that the program provided tenants with benefits amounting to about 5 percent of their income in the years 1965 and 1968. Additional evidence shows that tenant benefits declined rather sharply between the two years, and that in each of the years benefits were higher for older tenants, richer tenants, and white tenants than for their counterparts. Finally, evidence presented here suggests that, due to the in-kind nature of the program, the cost to landlords exceeded the benefits to tenants by about 75 percent.
